Time Series Forecasting Applications

Timothy Manning

BUS 461

Mirza Murtaza

June 18, 2012

Over the past five weeks of this course I have learned about many different modeling and forecasting techniques. I have learned, from this class that I have developed and am using a time series forecasting technique at Coca-Cola that enables me to plan our business more effectively. I can see the benefits this technique can present to a large number of companies that choose to utilize forecasting techniques, including Coca-Cola. I will attempt to identify some of the main issues and techniques associated with time series forecasting. I will also cover and reference some of the new things I have learned from my own research along with some things that I have learned from my classmates post. Most importantly, I will give some specific situations on how time series modeling can benefit the workplaces of companies that choose to use these techniques and how those who don’t can suffer from the lack of planning and foresight.

Time series forecasting techniques involve decision analyses which are often based off of past data that can help spot and identify trends and can enable a company to more accurately forecast and anticipate future needs of their business. A couple types of forecasting models are stationary and time series. The stationary forecast model represents, “the simplest time series forecasting model, it is one in which the mean value of the item being examined is assumed to be relatively constant, or stationary, over time.” (Lawrence, et al.,2002 p.390). On the other hand, the time series model considers patterns that may result from calendar, climate, or economic factors. Therefore the time series approach is good to use to help identify trends from seasonality or cyclical variations that will affect the overall model. There are two types of time series models which are additive and multiplicative. Our book says “the major difference between the two models is how one...